* mkldnn->onednn

* fused softplus op + kernel

* remove extra attributes

* add missing handler

* change var name

* move io

* fix ci bugs

* fix ci bugs

* fix py3 bugs

* fix example code

* fix example code

* fix text

* fix text

* deal with ci bugs

* perfect code according comment

* delete import batch

* Add gru qat int8 test

* Change place of model downloading

* Update paddle/fluid/inference/tests/api/CMakeLists.txt
Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com>

* Correct flags names and add description

---------
Co-authored-by: NSławomir Siwek <slawomir.siwek@intel.com>

* got rid of save_quant_model

* review changes

* remove with_data_parallel in unittest

* fix CI

* remove comment

* trigger CI

* revert part changes

* test_build_strategy_fusion_group_pass

* remove extra input of conv2d

* fix bug

* fix unittest bug

* adjust conv2d.pbtxt

* fix cpu_quantize_pass_tester

* revert use_addto of conv2d

* fix runtime attribute

* fix bug

* recover force_fp32_output in conv2d

* refine error info

* fix bug

* update config

* re-blacken python code

* temporarily disable date and diff_py_file

* skip a format